چكيده به لاتين
In recent years, a new efficient method of interference management has been revieded, known as interference alignment ( ). Using this method, for the -user interference channel with single antenna at each node, degrees of freedom ( ) or multiplexing gain ( ) is achievable, i.e. each user achieves half the . In this method, each receiver signal space is split into two equal-sized subspaces. Then the precoding matrices are designed such that at a specific receiver, all interference signals are completely located in one of the two subspaces, leave the other one free for the desired signal.
Achieving the upperbound of has been always a main goal to reach higher data rates. Using the traditional method of , proposed by Cadambe and Jafar, an infinite number of symbol extension is required to achieve an upperbound of , i.e., with finite dimensions, it can only gotten close to it. Therefore, since using the infinite number of dimensions is not practically possible, scheme with less number of channel extension and further total is more efficient. After the , Jafar's proposed ( ) design. The proposed a new structure for precoding matrices such that with less symbol extension, higher than the scheme is achieved.
In this thesis, by beam forming, a new efficient structure for precoding matrices is proposed which for the reduced channel extension, higher is achieved. The key idea of the new method is to use the combinatorics and number theories and in particular the principle of inclusion and exclusion, to design precoding subspaces. Obviously, the increasement of the will have a direct impact on the data rate performance. Furthermore, due to the significant reduction of the symbol extension, the size of the channel matrices is reduced, leads to the reduction of computational complexity of the whole system.
Keywords: Interference Alignment, degrees of freedom or multiplexing gain, Beam forming, channel extension.